We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
This uses this hack to generate seamless image right at the inference stage. Results might vary depending on the model and prompt.
Here is a few output from an extended version of the available example. The main difference is that I use an upscale step before running deep bump.
The nodes strikethroughs are just not documented yet so it's easier to spot empty wiki pages.
